1 | Author: Dimosthenis Sdrolias
Student Number: 11376260
Date of submission: 22-06-2017
Version: Final Master Thesis
Qualification: MSc. in Business Administration – Digital Business track
Institution: UvA
Supervisor: Ruben de Bliek
Positive WOM influence on consumers on pre-purchase
stages and the role of familiarity and trust: An
2 | Statement of Originality
This document is written by Dimosthenis Sdrolias who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.
The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
3 | Table of Contents
1.0 Introduction ……….………..Pages 4-8 2.0 Literature review………Pages 8-21 3.0 Data and method…….………...Pages 22-27 4.0 Analysis and results……….………...……...Pages 28-36 5.0 Discussion………..Pages 36-40 6.0 References………..Pages 41-47 7.0 Appendices……….………Pages 48-55
Abstract: This research is an experimental study on positive word-of-mouth influence at the pre-purchase stage of consumers in the FMCG industry. WOM is not something new, however digitization opened new horizons and created a complex environment for academics, practitioners and consumers. Mastering WOM activities can provide huge advantages to marketers now and in the future. This study examined 3 types of WOM namely, Blogs, Twitter and traditional WOM, across six different brands of the FMCG sector. The results indicate that consumers perceptions are indeed influenced by positive WOM and this influence depends on the level of trust on the different WOM platforms. The conclusions of this study may help marketers create more efficient WOM activities in the beginning of the customer journey, and will create breeding ground for further research to the extant theory.
4 | 1.0 Introduction
igital technologies advanced rapidly in the latest years. Consumers find themselves in an empowered position, in which they have a variety of tools in their possession, which often place them in the driver’s seat position in the customer journey (Edelman and Singer, 2015). New online media channels are available, creating a breeding ground for consumer communication and fast information dissemination (Cheung and Thadani, 2012). Online social communities and platforms have become essential sources of information for consumers, in order to facilitate their decisions and purchases (Cheung, Xiao and Liu, 2014).
Today, both companies and consumers are creators of content concerning a brand or a product. Marketers must adapt, by being alarmed and prepared to deal with any kind of situation, 24/7. However, this is not something new. Traditional word-of-mouth (from now on referred as tWOM) always existed and influenced consumer behaviour. A widely used definition of tWOM is the following. “Word-of-mouth is defined, as oral, informal, person-to-person communication between a perceived noncommercial communicator and a receiver regarding a brand, a product, an organization, or a service” (Higie, Feick, & Price, 1987).
Through digital technologies though, the ease with which consumers can share and deliver information and opinions is unprecedented. In addition, consumers have access to richer content and larger volume of information than they had via the tWOM (Kannan and Li, 2016). Academics and marketers refer to this concept as electronic-word-of-mouth (from now forth referred as eWOM). Hennig-Thurau et al (2004) define this concept “as any positive or negative statement made by potential, actual, or former customers about a product or company, which is made available to a multitude of people and institutions via the Internet.”
Yan et al. (2016) use 2 main categories of eWOM touchpoints, E-commerce touchpoints (referred as EC-eWOM) and social media touchpoints (referred as SM-eWOM). The first
D
5 | category is consisted by e-commerce websites that offer the possibility of online reviews. While the second one, is consisted mostly of consumer posts in social media platforms including microblogs, such as Twitter, Facebook and YouTube. Referred also as Social Networking Services (SNS), enable users to send and receive short messages, usually restricted in a specific number of characters. The message can take many forms, for example a picture (Instagram), or a video however, most social networking services, offer the option to add contacts, create a personal profile, share information and contact people connected to your profile. The above categorization of EC-eWOM and SM-eWOM is the most commonly used in the academic society. Those two types of eWOM have different characteristics. Some of the main differences between those two types are, firstly microblogs tend to involve more conversations and social interaction, not to mention that those posts happen in real time. While online reviews are being summaries, most of the times anonymous, bigger in size, reflecting opinions or experiences (Marchand, Hennig-Thurau and Wiertz, 2016).
In the category of SM-eWOM, blogs are also included, and have some distinct characteristics comparing to the above. Blogs (short for Web-logs), are websites in which the content is created by individuals, and also owned by them. Blogs may include any kind of content and usually refer to a specific topic (Berthon et al., 2012). Messages reflect bloggers’ opinions and may have a persuasion power over the readers’ mind (Jensen Schau and Gilly, 2003, Uribe, Buzeta and Velásquez, 2016). Blogs are a source of WOM but in many cases an advertising tool for marketers (Uribe, Buzeta and Velásquez, 2016). In this study, we will focus on the aspect of blogs as a WOM source.
All of these new channels that consumers use to get informed and disseminate messages to their social circle, in association with the tWOM, create a very complex and unknown environment for academics and marketers. The different characteristics and WOM channels used in the various industries add even more complexity and uncertainty. In many cases,
6 | benefits linked with WOM activities are neglected and remain unexploited. According to Forbes (2014) and American Marketing Association (AMA), 64% of marketers believe that word of mouth is the most effective form of marketing. Despite this fact, only 6% claimed that they have mastered WOM. The aim of his research is to shed light in specifics aspects of WOM that have yet not been explored by academics and may assist future marketers have a clearer view of WOM potential.
The extant complexity affected not only academics and marketers, but also the consumers as well. The large variety of different platforms and ways of digital communication that emerged with the rise of the internet, has created an overwhelming environment for consumers. Hence, consumers in order to reduce the uncertainty and evaluate information use different mechanisms. One mechanism that will be also investigated in this study is familiarity. Familiarity is the understanding of something by using past interactions and experiences (Luhmann, 1979, Gefen, 2000). Another mechanism sometimes related to familiarity that also is used by consumers is trust. Trust is referring to beliefs for upcoming actions of people or objects (Luhmann, 1979, Gefen, 2000). The two abovementioned mechanisms are often used by consumers and thus their role is of great interest for this study.
The importance of WOM nowadays, is highlighted by many authors. Triggering discussions and passion can make a big contribution to a brand (Forbes, 2014). Dias et al. (2016) from McKinsey & Company, found out that the companies that inspire their customers to generate WOM activities can achieve 30-40% more satisfaction than competitors. Studies conducted in the past also revealed that in general, WOM is a more trustworthy source of information than traditional paid media (Cheung and Thadani, 2012). This explains the huge number of literature existing in this field. The scope of those studies though, is very broad (Cheung and Thadani, 2012), and the results differ across industries. Thus, many gaps do exist concerning WOM literature.
7 | The purpose of this study is to address some of the existing gaps in literature that will be further elaborated below, by providing more specific conclusions concerning the Fast Moving Consumer Goods (FMCG) industry, and examine specific touchpoints of WOM and their influence on consumers. A typical FMCG product is “a low-priced item which is used rapidly with a single or limited number of consumptions (as opposed to consumer durable or consumer service).” (Baron et al., 1991, p.83). In this sector there are four main subcategories of products, home and personal care, food and beverage, tobacco and alcoholic drinks.
FMCG is an industry where the low price of the items and their low cost of production is resulting in a large influx of new products. Many of the products are similar to each other, hence the job of differentiation becomes even harder for marketers. Companies spend huge amounts of money in the traditional ways of marketing, despite the fact that they compete in an industry where WOM plays a very important role on consumers decisions. In this study the important influence of WOM will be examined in detail, taking into account the different WOM touchpoints, and their influence on consumers purchase intentions, the perceived quality concerning a product, and differences between the touchpoints and the consumers intention to share and disseminate WOM information. The WOM touchpoints that will be investigated are tWOM, mentions on blog posts, and mentions on Twitter. This study will be focused on the pre-purchase stage of the customer journey. The research method that will be used in order to investigate the abovementioned facts is, online experiment.
There are certain expectations and aspirations concerning the results of this study. To begin with, I believe that my research will contribute to the theory gaps concerning the understanding of WOM implication in the pre purchase level, where consumers seek and value information the most. The conclusions will refer to FMCG category but there is a possibility that some of the results may be applicable in a broader scope. By filling this gap on the theory marketers in the future will be able to better understand the power of WOM in the FMCG
8 | sector, and will eventually be more comfortable to use WOM activities along with the already well-established marketing activities. Furthermore, the results will reveal the comparison between Twitter, blogs and tWOM, importance at the pre purchase level. Thus in the future marketers will also be able to adjust their WOM activities according to the purchase level their customer may be. This will also assist in creating more personalized WOM activities that are aligned with each customer’s position in the customer journey. This research will also be of considerable use when considering which WOM touchpoints are more effective when a new product is launched. All consumers are by default on the pre purchase stage when a product is not already launched. As a result, those insights will give interesting information for marketers in order to finalize their strategy when there is a new product launch.
Hence, considering all of the above the research question of this study is stated as:
How blogs, Twitter and tWOM, influence consumers of FMCG products, at the pre-purchase level and what is the role of WOM touchpoint familiarity and trust?
2.0 Literature review
For many years word-of-mouth was considered to be a side effect of marketing activities. Still today, even though marketers try to develop a systematic approach (Peres, Shachar and Lovett, 2013), they seem concerned about eWOM marketing activities, and in many cases avoid them. Those concerns are derived mainly from the perception that WOM is still an immature and not of proven value concept, compared to traditional media and other more sophisticated concepts of marketing. Thus, marketers follow a more reluctant approach, not willing to question or change marketing strategies or tools that are well established for years (Bughin, Doogan and Vetvik, 2010). However the benefits from harnessing excellence in this field are profound. In traditional marketing activities all participants have access and use the same amount of information, making the incremental gain from excelling among competitors relatively small.
9 | On the other hand, few decide to seize the benefits from managing WOM, making its potential exponentially bigger (Bughin, Doogan and Vetvik, 2010). In another research by Trusov, Bucklin and Pauwels (2008), the results showed that WOM may enhance the impact of traditional marketing activities, when those activities aim to stimulate WOM.
As already pointed out by academics, online feedback mechanisms, as either EC-eWOM or SM-eWOM, can become a low cost channel for customer acquisition and retention, but can also assist in the fast-spreading of negative news and impressions (Dellarocas, 2003). With the huge technological advancements that took place in the latest years, it is a fact that the Web not only offers the possibility to disseminate information, but also to collect and interpret it (Dellarocas, 2003). Consumers’ online activity is now traceable, and provides a great opportunity to shed light in the consumer’s path to purchase (Srinivasan, Rutz and Pauwels, 2015).
A research by Jansen et al. (2009) examined microblogging as form of eWOM communication and its influence on branding. More specifically, the authors were focused on Twitter and they concluded that approximately 19% of total tweets mention a brand or an organization. This fact highlights the importance of microblogging for marketers and proves that there is plenty of space for marketing activities in this path. In addition, around 20% of those mentions were expressing a sentiment or opinion towards a brand, a product or an organization. Generally, the authors illustrated the impact of social communication channels on brand awareness and brand image. By reading this article the questions that emerge are, how do those ‘tweets’ or posts may potentially influence consumers that are considering to purchase a product for the first time? Are these posts important to trigger consumers into buying something they had not acquired before? It is also indicated in this study that the impact on brands is different in different industry sectors.
10 | 2.1 Research on pre-purchase stage
Many researches have been made on WOM but none of them dug deep into specific stages of the customer journey. Consider Edelman and Singer customer journey (Figure 1). The customer journey has specific stages (consider, evaluate, buy, enjoy, advocate, bond) from which consumers go through. This is a typical path when a new product is introduced to consumers mindset. Considering FMCG sector, which is a sector that has a large influx of new products, and knowing the great importance attributed to WOM by consumers in order to get informed, the question that arises is how important are the different WOM touchpoints, that consumers use to acquire information, and what is the influence of those touchpoints on consumers perceptions, like the perceived quality of a certain product or their purchase intentions.
Figure 1: The loyalty loop (Adapted from Edelman and Singer 2010)
Why though focus on specific stages of the journey and more specific in the pre-purchase stage? To begin with, consumers have different attitude, knowledge, expectations, amount of information, even choose different channels to get informed. The influence, that WOM and also other marketing activities have, is not the same for consumers who are in different stages of the journey. The challenge, as presented by Edelman and Singer (2010), is to be able to lock-in consumers in the loyalty loop, meaning a shorter journey that does not include consideration and evaluation steps. However, in order to succeed in the above challenge companies and
11 | academics must comprehend what drives and influence consumers that go through consideration to the purchase steps in terms of WOM. By knowing this, companies will be able to retain the customers throughout their journey and be able to put them in the loyalty loop. Despite the huge number of different views by authors concerning the customer journey, most of the views are having common steps in the process. For the completion of this research, the above customer journey introduced by Edelman and Singer (2010) will be use as it is quite detailed and assists in illustrating the importance of the pre-purchase stages in the journey.
Furthermore, Bughin, Doogan and Vetvik (2010) indicate that WOM is an important influential factor for 20-50% of all purchase decisions. Its influence though is greater when consumers purchase products for the first time, and due to digitization, this influence is expected to grow even bigger. The authors also indicate that the influence of WOM in the stages of consideration and evaluation are higher comparing to the latest stages of the consumer journey. Thus, future research should focus on giving answers and shedding light to WOM and its influence on consumers at this stages of their journeys.
Kim and Hanssens (2017) was the most relevant study found, addressing a similar topic from a different perspective. In their research, they examined how the consumers’ interest is influenced by advertisement and WOM activities during the pre-launch period. They concluded that in order to increase the amount of consumer searches for a product, blogging is a better option than advertising. In addition to that they found that blogging has more permanent effects than advertising. The above findings of this article illustrate the importance of blogging as an eWOM touchpoint, especially when consumers do not know any information about a product and have not yet acquired it. This article is undoubtedly insightful, however other researches should also examine other WOM touchpoints and other product categories, than the experience goods products investigated by the authors.
12 | 2.2 Research on FMCG
While traditional and electronic WOM have been extensively researched, most of the researches are focusing on specific product categories or industries. Numerous articles can be found concerning, electronical goods, tourism or more general hospitality services, however there is not much research done concerning FMCG products. Despite this lack of attention, FMCG sector appears to be a really important and healthy leaver of the economy. According to PWC, “Global Top 100 Companies by market capitalization” report, consumer goods industry ranked 3rd in market cap (2,612 bn USD, see Appendix), while 2 FMCG companies are listed among top 20 companies with the highest market cap. Nielsen also estimated that, FMCG sector nominal growth in Europe in Q2 2016 versus Q2 2015 was 0.8% (0.7% price increase and 0.1% volume increase, see Appendix) which is the lowest since 2008, however this fact does not undermine the importance of this sector in Europe’s economy.
Marketing science and more specifically consumer marketing, is evolving rapidly the latest years. With the internet boom and the ease of data and information dissemination, there is a shift in marketing activities with increasing emphasis on interaction, relationships, loyalty and retention, instead of traditional transactional marketing mix management (Leahy, 2011). Despite this trend in the FMCG marketing activities, little is known about the role and influence of eWOM and tWOM in this sector. More research is needed to comprehend the WOM influence on consumers in the purchase path, in order to use more effective consumer marketing activities towards building of relationships, and more effectively guide the consumer through the customer journey.
But, why the results of electronical goods or tourism cannot be relevant for FMCG? As it is derived from the definition, FMCG products are not very expensive, and in most of the times respond to utilitarian needs of the consumers. Those products are usually non-durable and sold
13 | quickly. Those facts indicate that the consumer has low-involvement when making decisions about buying a product from this category, even though there are exceptions when products have a strong brand loyalty. On the contrary electronical goods may be either utilitarian or hedonic, but their high cost, force the consumer to be more involved in the purchase process. Concerning tourism activities, they are bought mostly to satisfy hedonic needs and the cost in this sector is also higher. The above characteristics indicate why FMCG should be examined separately and why the results of other industries may not apply for this sector.
FMCG sector has distinct characteristics from other industries. The wide range of the products and the variety of the needs they satisfy make it difficult for researchers to make concrete conclusions. This sector is known also for the fierce competition between the numerous companies that operate in this field, the intense marketing efforts required to reach the consumer, but also for the high launch cost required to introduce a new product. The result of the above fact is that FMCG industry has the highest share of advertisement spending according to Nielsen (2016) (See Appendix D). Hence, for the above reasons, it is important to investigate the aspect of WOM to fill existing gaps in the literature and also to assist marketers with their decisions and actions in the industry. On top of the above, as Accenture Consulting (2016) also highlights (see Appendix B), consumer and customer knowledge is one of the five keys, for successful growth in the consumer goods industry, and this study will assist in broaden existing knowledge on this direction.
One research by Wu and Wang (2011) investigated the effects of message appeal and source credibility on brand attitude, which is very interesting. The authors used two fictional products, one from the electronic goods category and one from the FMCG category, to conduct their research. They found a direct and positive relationship between message source credibility and brand trust, brand affection, purchase intention, brand attitude, and support. They also found that product involvement would moderate the relationship between eWOM message
14 | appeal and brand attitude. However this research, did not take into account the different eWOM touchpoints that can play a critical role and potentially have different influence on consumers’ minds. Also as the authors suggested, real brands can be investigated in the future in order to have a better reflection on the consumer reality.
2.3 Comparison between eWOM and tWOM
Despite the wide literature existing, concerning WOM, still we lack of understanding how tWOM and eWOM differ from each other (Berger & Iyengar, 2013). There is little research to be found, in which comparison between those two forms is made. A relevant article is that of Meuter, McCabe and Curran (2013), which investigated the influential value of eWOM and interpersonal WOM. The authors found that interpersonal (traditional) word of mouth had bigger influence on behavioral intentions than eWOM, as it was perceived as a more trusted source of information. Furthermore, this research also revealed that customers valued the information and opinions gained via Facebook or independent review sites higher than company owned website or Facebook page. Despite these interesting findings, this research focused on the restaurant industry, thus we cannot apply the findings in FMCG industry as there are fundamental differences between them. Restaurant industry is a service industry in which the way people use WOM is totally different. To illustrate this point there is even a social media platform, Foursquare that is based on reviews of restaurants and other places in general. The restaurants most of the times satisfy hedonic needs and the cost is relatively higher than that of an FMCG product.
Peres, Shachar and Lovett (2013) also examined traditional versus electronic WOM and their impact on brands in the US. Their findings indicate that new brands are more discussed offline while, premium brands are discussed more online. However, in this study the authors use the term ‘brand’ in a broad sense, even though the conclusions may refer to specific
15 | product categories. In their brand set, there are products, services or stores from numerous industries. As the authors also highlight in the limitation sector, future research can study the different effects of specific channels and touchpoints of WOM. Last but not least, individual and consumer characteristics were not taken into account.
2.4 Message source credibility - Trustworthiness
People inevitably interact with other individuals in their everyday lives. This interaction, given the diversity of the human minds and personalities, is creating a very complex environment for individuals. As a result people seek methods to reduce this complexity and one of the ways to accomplish this is trust (Gefen, 2000). Trustworthiness in communication is described as the level of confidence and acceptance of the receiver perception, concerning the sender and the message. Individuals use trustworthiness not only in their interactions with individuals but also in their interactions with platforms. When people are subjected to WOM messages, the trustworthiness of the platform the message was on may actually play an important role in the influence this message will have on consumers. Chen et al. (2014) investigated in their research the role of trust in e-commerce platforms and found out that trust in a platform positively influences trust in sellers and as a consequence the customers’ purchase intentions are positively influenced.
While academic researches pointed out that credible sources have more persuasive power on consumers than low credible sources, Ohanian (1990) argues that highly credible sources are not always more effective. In addition, Ohanian (1990) uses 3 dimensions to measure the source credibility, which are expertise, attractiveness and trustworthiness. Not all three dimensions of credibility are applicable for Blogs and Twitter. Expertise and Attractiveness attributes are not valid for those platforms as they are involving the actual person that posts and not the platform itself. Thus, we will only investigate the trustworthiness aspect of those
16 | platforms, and concerning the recommendation by friends this will give as an indication of how trustworthy in general consumers think of tWOM.
2.5 WOM touchpoint familiarity
Another method that people use to reduce complexity is familiarity. There are some studies being done that investigate the role of brand familiarity on consumers which is often associated to brand awareness. Johnson and Russo (1984) pointed out that brand familiarity enhances the ability to encode and remember new information. In this research brand familiarity will be absent or low as consumers will be at the pre purchase level. Even though many more have examined the role of brand familiarity, the role of WOM familiarity in consumers’ minds is still vague and not studied by academics. The familiarity that consumers have with the different touchpoints that are interacting with, is potentially moderating the influence that WOM has on consumer purchase intention, perceived quality of products and the message share probability. In other words, the relationship between WOM and consumer perceptions depends on the level of familiarity towards WOM platforms. Practically, this means that, the more familiar a person is with a specific WOM touchpoint, the more influenced he will be concerning his perceptions and future intentions and vice versa. Hence, in the context of this study we will investigate this possible moderation effect that the different WOM familiarity levels have in the influence on consumers. The results will actually reveal useful information for both practitioners and academics, who will be able to understand the role of WOM familiarity in the messages influence on consumers perceptions, and reveal insights how to create more targeted and effective campaigns.
2.6 Research Gap
First of all, in the vast majority of the articles reviewed, it is mentioned that the results cannot be generalized to other industries. This is a common conclusion of the academic articles,
17 | and is based on the different characteristics of the various product categories, the different type of needs they satisfy and the level of the customer involvement. That is the reason why most authors focus on specific product categories. However, the research done on the FMCG product category is scarce. Not only this sector differs in the above characteristics as indicated above in this study, but also the touchpoints and channels used by consumers differ significantly. To illustrate this point EC-eWOM is not common in the FMCG sector and consumers are using more SM-eWOM and tWOM to gain information and assist their purchases. As a result, I believe that many insights are yet to be discovered concerning FMCG products consumers and WOM influence on their decisions.
In addition to that, I found that there are only a few experimental studies in this field. Experimental studies can bring additional insights and better understanding of the consumer’s mind. It is also, a better way to investigate causal relationships. Furthermore, the comparison between traditional and electronical WOM is still not fully investigated. Most of the papers that compare these two concepts do not compare specific channels/touchpoints but they instead use a more abstract approach.
Last but not least, there is no research yet conducted in order to investigate the influence of WOM to the consumers’ behavior at the pre-purchase step. When consumers have not yet used a product and do not have prior experience about its usage. The motivation behind this choice derives from the fact that people use WOM more when they are at this stage of purchase. It is also indicated by the authors that WOM is more important for the consumers in this stage comparing to post-purchase stages.
18 | 2.7 Touchpoint selection
2.7.1 Twitter
The decision to choose Twitter as a social media platform was based on its characteristics. A research by Aladwani (2015) revealed that one of the three main reasons that drive individuals to use Twitter is WOM. Furthermore, Twitter impact on individuals is concerning relationships, knowledge and satisfaction (Aladwani, 2015). The knowledge impact of Twitter is higher compared to other social media platforms. Another important fact that influenced my decision is that according to Rui, Liu and Whinston (2013), Twitter has a bigger awareness effect than other social media platforms and eWOM touchpoints in general. Awareness is key in order to reach consumers and trigger them into purchase. Lastly, “following” a person on Twitter comparing to “friending” a person on Facebook, does not require a deep connection between the users. Subsequently, in Twitter users many times “follow” other accounts for content purposes. Thus, all of the abovementioned facts make Twitter ideal choice for my research.
2.7.2 Blogs
The decision to include blogs is derived from the nature and characteristics of the FMCG category. As mentioned already in this proposal, due to the specific characteristics of the products, there are also differences in the touchpoints consumer use to get informed and facilitate their decisions. Except from social media platforms consumers use blogs a lot to get informed and do not use e-commerce review sites or other means of online reviews. A digital influence report created by Technorati Media (2013) revealed that blogs were the third most influential digital source of information, outmatched only by retail sites and brand sites. That explains also why today more and more marketers use blogs as an advertisement tool (Uribe, Buzeta and Velásquez, 2016) In addition, like Twitter, blogs also have an awareness effect, as
19 | many new products are tried by bloggers and people that have not tried them before can collect some initial impressions and information about them. Lastly, as indicated by Kim and Hanssens (2017), blogs are a good tool for marketers at the pre-launch stage when consumers are by default on their pre-purchase stage of the customer journey. Thus, since my research will be focused on FMCG it is essential to investigate the impact of blogs on consumers.
2.7.3 tWOM
According to Eisingerich et al. (2015) the social risk on eWOM is considered as higher than in tWOM. The social risk is even lower when someone is sharing information with a friend or family. In this research I chose tWOM in order to examine its influence on consumers compared to blogs and tweets, considering its lower social risk. Furthermore, tWOM is less controllable and measurable compared to eWOM, thus it would be insightful to examine a comparison between them.
2.8 Hypotheses development
In order to answer the main research question the following hypotheses must be examined. In this study only consumers that are in the pre-purchase level of their journey will be examined, thus all the bellow hypotheses are by default on the pre-purchase level of consumers, meaning that consumers have not yet acquired the products, and have no experience on their usability and performance.
2.8.1 tWOM versus eWOM:
Meuter, McCabe and Curran (2013), on their research on the restaurant industry made a comparison between the influence of eWOM and interpersonal WOM concluding that interpersonal WOM did have a greater impact on consumers behavioral intentions. Peres, Shachar and Lovett (2013) also tried to prove the same point from a more general point of view.
20 | Both studies used the term eWOM and tWOM in a more abstract way without taking into account specific channels. However, in this study a more detailed approach will be adopted as specific channels’ influence on consumers will be examined, and based on the above findings the following hypotheses for FMCG category, are expected to be proved from this experiment study.
H1a: tWOM influence on consumers’ purchase intention is higher than the influence of Twitter or blog posts.
H1b: tWOM influence on consumers’ perceived quality concerning a product, is higher than Twitter or blog posts.
H1c: Consumers are more willing to repost or share a message from friends or family than a message on Twitter or on a blog post.
2.8.2 WOM touchpoint familiarity
As explained previously in this paper, the level of consumers WOM touchpoint familiarity, and its role is something that is still not investigated in the academic literature. I believe that the WOM touchpoint familiarity will actually moderate the influence that the WOM messages have on consumers’ perceptions, meaning that the higher the familiarity with the different WOM touchpoints the higher the influence. Even though there is no academic literature referring to this type of familiarity, there are other researches that imply that different types of familiarity play important roles. Johnson and Russo (1984) study shows that brand familiarity is crucial enhances the ability to remember information and also affects the influence of advertisements concerning products. Hence, also based on their findings I can create the following hypothesis.
H2a: WOM touchpoint familiarity moderates the influence of the WOM messages on consumers purchase intention.
21 | H2b: WOM touchpoint familiarity moderates the influence of the WOM messages on consumers perceived quality about products.
H2c: WOM touchpoint familiarity moderates consumer intention of reposting or sharing the WOM messages that they were exposed to.
2.8.3 WOM touchpoint trustworthiness:
In this research we study the different WOM touchpoints and their influence on consumers’ perceptions in the pre-purchase stage. Something we must also take into account is the role of the platform/touchpoint trustworthiness for consumers. I support the view that high trustworthiness of the WOM touchpoint will have a positive moderating effect on consumers’ perceptions. Szulanski, Cappetta and Jensen (2004) claim that the perceived trustworthiness of the source may influence the behavior of the recipient, while Ohanian (1990) argues that highly credible sources are not always more effective, thus an answer is yet to be given on whether trustworthiness of the different WOM touchpoints has indeed a moderating role in the consumers perceptions. Chen et al. (2014) that investigated the role of trust on e-commerce platforms, found a positive influence between trust and purchase intentions, hence considering the above theory I state another hypothesis on the moderating role of trust in WOM and its influence on consumers as follows.
H3a: WOM touchpoint trustworthiness moderates the influence of the WOM messages on consumers purchase intention.
H3b: WOM touchpoint trustworthiness moderates the influence of the WOM messages on consumers perceived quality about products.
H3c: WOM touchpoint trustworthiness moderates consumer intention of reposting or sharing the WOM messages that they were exposed to.
22 | 3.0 Data & method
3.1 Research design and methodology
The research method chosen in order to conduct this study is online experiment. For this study I will deliberately expose a group of people to a treatment and then observe their response. Different people will be exposed to different treatments. More specifically participants of this study will be exposed to different positive WOM mentions, concerning six different brands of the FMCG sector. Then, I will investigate if the influence of WOM on consumers purchase intentions, perceived quality and intention to share, is moderated by how familiar the participants are with the WOM touchpoint and by their perception on the level of trustworthiness of those different touchpoints. I have chosen a mixed design for this online experiment. The brand variable will be examined as within subject while the WOM touchpoints will be examined as between subjects. The tool that was used in order to create the online questionnaire and to conduct this online experiment is Qualtrics.
23 | 3.2 Operationalization of variables and measurement:
In this experiment we want to investigate the influence of WOM different touchpoints on consumers of FMCG products in the pre-purchase level. In order to do so, I will use two independent variables, and through their manipulation I will measure the impact on three dependent variables.
3.2.1 Independent variables
To begin with the independent variables, two of them were selected for the purpose of this experiment. Firstly, different WOM touchpoint mentions will be examined. Blog post messages, Tweets and recommendation by friends are the three different channels that are going to be investigated. The WOM message will be the same across the different alternatives for each brand in order to eliminate any text influence on consumers. A realistic message will be created that will be similar to actual mentions. In additions all messages are up to 140 characters which is the limit for messages on Twitter. In the Appendix sector an example of a message for each WOM touchpoint can be found. The different WOM messages were created in such a way in order to be perceived as real messages. Blogs were carefully used and were relevant to the products. For instance for Calve product (mayonnaise) a food blog was used. Same logic was followed for Tweets in terms of the source of the message. Lastly concerning the tWOM I used a story in order for the respondents to better be able to think and empathize with the concept and the message.
The second independent variable will be brand. In this study, six different brands across the FMCG industry will be carefully selected in order to perform this experiment. Consumers that will participate in this experiment are at the pre-purchase stage, thus, the brands selected are not globally known or iconic brands, instead, less known brands were selected to better serve the purpose of this study. Consumers will be randomly assigned to one WOM channel
24 | message per brand. The brands that are tested are carefully selected from the different FMCG categories. There are, one brand for food category, one for personal care, one for home care, one alcoholic drink and two from the beverages category. The aim of the usage of the different brands is to create a sample of brands representative for all FMCG categories (excluding Tobacco). In addition, due to the extensive volume of different FMCG product categories the six brands will reveal if any of the results is brand related or it is the same across all brands. A list of the selected brands of this study can be found at the Appendices sector.
3.2.2 Dependent variables
Concerning the dependent variables, two were selected in order to measure the influence of WOM on consumers. The first dependent variable that will be investigated is purchase intention. “Purchase intentions are an individual‘s conscious plan to make an effort to purchase a brand” (Spears and Sing, 2004). Even though purchase intention is actually a perception and not an action of the consumer, it is of great importance, because a measurement of the influence on consumer’s purchase intention will assist academics and marketers to comprehend the consumer behavior, and create more effective marketing campaigns (Puth, Mostert and Ewing, 1999). In addition Dias et al. (2016) also indicate that in order to create distinctive customer journeys for your consumers, perceptions are one of the factors that matters the most in this endeavor.
The second dependent variable selected is customer’s perceived quality. Perceived Quality is defined as “the customer’s perception of the overall quality or superiority of a product or service with respect to its intended purpose, relative to alternatives” (Zeithaml, 1988, Aaker, 1996). As discussed by Aaker (1996), perceived quality may contribute in many ways for the brand, one of them which is also suitable for this research, is that perceived quality provides with a reason-to-buy. Thus, considering that this study is focused on the consumers’
pre-25 | purchase level, where there is no prior experience about the brand, this variable can provide with additional insights on how people are influenced by WOM, and if stimuli from WOM may further motivate consumers to make a purchase.
The third dependent variable will be also examined in this experiment, the intention to repost/share the information by the respondents. In this study the results will reveal the potential influence of WOM in consumer’s purchase intention and perceived quality about FMCG products, however another important factor is if consumers are actually willing to disseminate that information to others as well. The results will provide insights in the existing literature concerning which channels of WOM consumers are most likely to share in their social circle, and also assist marketers by revealing which WOM channels they should focus on, in order to faster spread information to consumers especially when dealing with new products. A 7 point Likert scale was used to measure all of the above dependent variables and all of the variables were single item constructs.
3.2.3 Moderators
In the context of this research, I will also examine if other constructs moderate the effect of the WOM messages to the consumers’ perceptions. The first construct that will be investigated is the familiarity level of the consumers towards the three different types of WOM. Johnson and Russo (1984) pointed out that brand familiarity enhances the ability to encode and remember new information, hence I will investigate if this is also the case for the FMCG category and for WOM messages. In order to capture the familiarity level of the consumers with the different WOM touchpoints I will use a 1-7 Likert scale (1= Not familiar at all, 7= Extremely familiar), similarly to other studies as that of Qureshi et al. (2009), and will be a single item construct.
26 | The second construct that will be examined concerning its moderating effect is trustworthiness of the different WOM platforms. Trustworthiness, is a construct widely examined in e-commerce platforms and researches reveal that it influences consumers decision making and enhances the chance of purchase. Thus, I believe that trustworthiness of WOM touchpoints will play a moderating role for the messages derived from those sources. In order to capture the trustworthiness of information deriving from different platforms, a 7 point Likert scale (1= Not at all trustworthy, 7= Extremely trustworthy) will be used similarly to other studies like the one of Pan and Chiou (2011), and will be a single item construct.
3.3 Procedure & Sample
The experiment was conducted via an online questionnaire, and the fieldwork duration was approximately 2 weeks at the end of April 2017. The online questionnaire was created on Qualtrics and required approximately 8 to 10 minutes to be completed. The answers that were initially collected were from 205 participants. Before the starting the analysis various checks were made in order to assure that the answers collected were valid. First of all, there were a few participants in the sample that did not answer all the questions of the online experiment, hence they were removed. Secondly, I viewed the timings that participants needed to complete the survey, and deleted the participants that filled the questionnaire in less than 5 minutes, as the time was too less in order to read and respond properly to all the questions. A manipulation check was also included in the online questionnaire in order to reassure that participants were focused while responding. The 93,2% of the participants responded correctly at the manipulation check question, thus a review of the rest participants that failed to answer correctly was made in order to clean the data from peculiar answers. After all of the above checks the final sample of this study consisted of 162 participants.
27 | The experiment sample was consisted of 41.4% (67) male participants and 58.6% (95) female. The majority of the respondents (53.1%) were at the age group of 25-34 years old while 25.4% of the participants were between 18-24 years old. Concerning the educational background of the participants, Bachelor and Master studies graduates were the 88.3% of the sample. Lastly, in terms of nationality the vast majority of the sample was from Greece 66.7%, the second largest group was Dutch 10.5%, while the rest was from a variety of 22 other countries.1
At this experiment 6 different products were investigated. In order to check if the participants were at the pre-purchase stage of the customer journey there was a question in each product asking if participants ever purchased those products. Participants that answered ‘yes’ had their answers removed from the sample for the specific product. Thus, in the sample only those who did not ever purchase those products were included.
The structure of the experiment was as follows. Firstly, participants were asked for their initial purchase intention and perceived quality concerning the six brands (pre-treatment variables), and then they continued in the WOM treatment that was randomly either a blog post, or a Tweet or a message from a friend (tWOM). Then, after participants were exposed to the treatment message, they were asked once more about their purchase intention concerning those brands, the perceived quality and the probability of sharing the message they were expose to (post-treatment variables). After the randomized treatment, participants were exposed to the manipulation check question and then, were asked about the level of familiarity and trust they have concerning the 3 abovementioned WOM touchpoints. Lastly participants were answering the demographic questions before they finish the online experiment.
1 For more detailed information refer to Appendix G
2 For related information you may visit The Washington Post and TechCrunch relative articles in the below links
28 | 4.0 Analysis and results
Before starting analyzing the data some adjustments needed to be made. To begin with, calculation of the post treatment purchase intention, perceived quality and sharing probability for the 3 different WOM touchpoints. Secondly, the initial purchase intention and perceived quality needed to be calculated. In order to do this, I created initial purchase intention and perceived quality per WOM touchpoint. In other words, for instance initial purchase intention for blogs were counting only those answers for the products that participants were exposed to blog treatment, initial purchase intention for Twitter was counting the answers for the products that participants were exposed to Twitter treatment and so forth. Those computations were made for each participant, as the treatments were randomly appointed to products. To sum up, there were for each WOM touchpoint, initial purchase intention and perceived quality variables (before treatment) and post treatment purchase intention and perceived quality variables. Table 1 below shows the differences in the means of purchase intention and perceived quality between pre and post treatment and poses an initial impression of the influence that each WOM touchpoint had on consumers perceptions.
Table 1: Pre/Post treatment Mean comparison
PURCHASE INTENTION Pre- treatment Mean Post-treatment Mean
Blog 3.970 4.531
Twitter 4.007 4.444
tWOM 3.796 4.644
PERCEIVED QUALITY Pre- treatment Mean Post-treatment Mean
Blog 4.786 5.100
Twitter 4.763 5.074
tWOM 4.579 5.177
In order to continue with the analysis “Delta” variables were created. Delta variables were for each WOM touchpoint the computation of “Post treatment purchase intention” minus
29 | “Initial purchase intention” . The same computations were made for perceived quality variables. Higher levels therefore indicate a greater increase in purchase intention or perceived quality due to word-of- mouth messages shown in the experiment. The following table consists an overview of the means and standard deviations of the Delta variables.
Table 2: Delta variables Means & Std Deviation
PURCHASE INTENTION Mean Std. Deviation
Delta Blog 0.561 1.329
Delta Twitter 0.443 1.255
Delta tWOM 0.848 1.308
PERCEIVED QUALITY Mean Std. Deviation
Delta Blog 0.314 1.081
Delta Twitter 0.310 1.067
Delta tWOM 0.597 1.049
4.1 Hypothesis testing
Since there are variables that measured before and after a treatment the analysis that will be used is general linear model, ANOVA repeated measures analysis. To begin with, in order to examine if tWOM impact on the dependent variables was bigger compared to blogs and Twitter, nine different ANOVA repeated measures analyses were conducted. The results of the analysis (Table 3) show that tWOM influence is significantly higher than that of blogs or Twitter, at a 10% significance level. These results are consistent across the 3 different dependent variables therefore, subsequently for the 3 different sub-hypotheses. As a result of the above analysis shown in Table 1, H1a, H1b and H1c are supported, which means that indeed as expected, tWOM has bigger influence on consumers compared to blogs or Twitter. The above result indicate that consumers at the pre-purchase level that are exposed to positive WOM concerning a product have higher purchase intention and perceived quality towards that product than
30 | before. This increase is higher when consumers come across tWOM, for example a message from a friend, compared to blog posts or tweets.
Table 3: H1 testing results, F value, P value and df.
F value P value df Purchase Intention:
Delta Blog vs Delta tWOM 3.051 .084 1
Delta Twitter vs Delta tWOM 8.132 .005 1
Delta Twitter vs Delta Blog .370 .544 1
Perceived Quality:
Delta Blog vs Delta tWOM 4.849 .030 1
Delta Twitter vs Delta tWOM 10.965 .001 1
Delta Twitter vs Delta Blog .056 .813 1
Share probability:
Delta Blog vs Delta tWOM 6.829 .010 1
Delta Twitter vs Delta tWOM 24.432 .000 1
Delta Twitter vs Delta Blog 2.438 .121 1
Another important and interesting output is that in contrast to the above, difference in the influence of Twitter versus blogs were not significant across the 3 dependent variables. In order to further investigate the above outcome, a correlation analysis was made in order to examine if blogs and Twitter post treatment reactions are strongly correlated. Results show that there is a significant but weak correlation between the 2 WOM touchpoints across the purchase intention and perceived quality, but a strong correlation for share probability. For purchase intention the Pearson correlation was .282 (p=.002), for perceived quality Pearson correlation was .322 (p=0.001) and for share probability Pearson correlation .708 (p=0.000). This fact means that,
31 | when it comes to sharing a message, the probability to share will be similar irrespectively of the eWOM platform that consumers got this message from, in our case blogs or Twitter.
In order to test the second Hypothesis a similar logic was implemented for the moderation variable of familiarity. The familiarity variable was measured in a 7 point Likert scale for each WOM touchpoint. A similar procedure was adopted for this variable meaning that “Delta” variables were created by computing “tWOM Familiarity” minus “Blog familiarity”, “tWOM familiarity” minus “Twitter familiarity”, and “Blog familiarity” minus “Twitter familiarity”. Then the different Delta familiarities were grouped as follows. When the number was bigger than 1.5 points (7-point Likert scale), was coded as 0 and all other values were coded as 1. With this grouping we consider two groups of people, those who considered themselves to be a lot more familiar with tWOM than blogs or Twitter and those that do not consider themselves as more familiar with tWOM comparing to the other two eWOM touchpoints.
After the above computations, three Repeated Measures ANOVA were conducted in order to test the familiarity moderation effect. The results for both Delta Blog versus Delta tWOM and Delta Twitter vs Delta tWOM concerning familiarity moderation were not significant for each of the three dependent variables. More specific for purchase intention Delta Blog vs Delta tWOM results were (F=1.072, p=.314, df=1) and for Delta Twitter versus Delta tWOM (F=.144 ,p=.703 , df=1). The same results for perceived quality, Delta Blog vs Delta tWOM (F=2.075 ,p=.153 , df=1) and Delta Twitter versus Delta tWOM (F=.018 ,p=.894 , df=1).
Despite the above findings, when looking at the descriptive statistics and the Mean differences we can see difference between the groups. More specific, for instance at purchase intention variables the Delta Blogs Means in the two different groups of familiarity is .556 and .602 for Twitter .431 and .499 and for tWOM .719 and .923. In addition to that, from the correlation analysis made at H1 testing, it seems like Twitter and blogs have similar influence
32 | to consumers. Hence, in order to reassure that indeed there is no moderation effect another analysis will be conducted comparing eWOM (Blogs & Twitter merged) with tWOM the summary of the results is at Table 4. The results indicate that the H2 (2a, 2b and 2c) is not supported as neither of the influence on the depended variables seems to be moderated by familiarity levels of the different platforms. This fact means that the influence from positive WOM is not depended to the level of familiarity consumers have with the different WOM touchpoints namely, blogs, Twitter or tWOM.
Table 4: H2 testing, F values, P values and df.
F value P value df Purchase Intention:
Delta eWOM vs Delta tWOM 1.329 .251 1
Perceived Quality:
Delta eWOM vs Delta tWOM 1.070 .303 1
Share Probability:
Delta eWOM vs Delta tWOM .887 .348 1
The same procedure was followed in order to test the third hypothesis concerning the moderation effect of trustworthiness of the information derived from the 3 different WOM touchpoints. Trustworthiness of the WOM touchpoints was measured by a 7-point Likert scale, one for each touchpoint. Therefore, ”Delta trust” variables were created, that were consisted of Trust tWOM minus Trust Blogs, Trust tWOM minus Trust Twitter and Trust Blog minus Trust Twitter. The outcome of this subtraction was then grouped as follows. When the number was bigger than 1.5 points (7-point Likert scale), was coded as 0 and all other values were coded as 1. With this grouping we consider two groups of people, those who trust a lot more tWOM than blogs or Twitter and those that do not consider those WOM touchpoints as very different
33 | concerning the trustworthiness of information. The analysis to test this Hypothesis was also general linear model repeated measures ANOVA. The different types of WOM were analyzed with each other and the moderation effect of trust was investigated. The results show that both Blog vs tWOM (F=.615 ,p=.434 ,df=1) and Twitter vs tWOM effect was not significant (F=.274 ,p=.602 ,df=1) for purchase intention and trust moderation. Similar results for perceived quality variables, more specifically, blog vs tWOM (F=2.691 ,p=.088 ,df=1) , Twitter vs tWOM (F=.112 ,p=.739 ,df=1) and for share probability variables where Blog vs tWOM (F=0.072 ,p=.790 ,df=1) and Twitter vs tWOM (F=1.310 ,p=.255 ,df=1).
Despite the above results when viewing the descriptive statistics and the means of blog Twitter and tWOM across the two trust groups that were created, a similar trend can be spotted as the one in H2. It seems that there is a difference in Delta tWOM purchase intention Means (Mean0=.750 Mean1=.945) among the two groups and also smaller differences in the means of Delta Blog (Mean0=.792, Mean1=.825) and Twitter (Mean0=.460, Mean1=.497) purchase intention across the 2 groups. Also taking also into account the correlation analysis that was made in Hypothesis 1, before deciding whether to support this Hypothesis another final analysis will be conducted.
The same procedure will be followed but this time blogs and Twitter will be grouped into eWOM. Therefore, the moderation effect of trustworthiness will be tested across eWOM and tWOM. The results are summarized in the bellow Table 5. After the bellow Repeated measures ANOVA analysis I partially support the third Hypothesis. H3a and H3b are supported while H3c is not supported. From the results it seems that purchase intention and perceived quality are moderated by the trustworthiness level of the different WOM touchpoints (in a 10% level of significance), while the share probability and trust moderation effect is not supported. This means that the impact of positive WOM on consumers purchase intention and perceived quality is depending on the level of trust consumers have towards tWOM and eWOM. Results are
34 | summarized in table 5 and in figures 3 and 4 below. In figures 3 and 4 the line 0 depicts the behavior in terms of purchase intention (Figure 3) and perceived quality (Figure 4), for those participants who trusted tWOM and eWOM approximately the same (difference was less than 1.5 points in a 7-point likert scale). The 1 line shows behavior of those participants who trusted one form of WOM a lot more than the other (difference was more than 1.5 points in a 7-point Likert scale).
Table 5: H3 testing , F value P value and df.
F value P value df
Purchase Intention:
Delta eWOM vs Delta tWOM 3.208 .076 1
Perceived Quality:
Delta eWOM vs Delta tWOM 6.669 .011 1
Share probability:
35 | Figure 3: Moderation effect of Trust in Purchase Intention before and after WOM messages treatment.
0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9 1 1 2
0 (No/Small difference in Trust) 1 (Big difference in Trust)
Est im at ed Ma rgi nal Me ans Purchase Intention
36 | Figure 4: Moderation effect of Trust in Perceived Quality before and after WOM messages treatment.
5.0 Discussion
Digital technologies advancement of the latest years has created a different landscape for consumers in many industries. FMCG is a large industry that is inevitably affected by digital technologies. Despite this fact, there is a wide belief in the sector that still traditional WOM is the touchpoint that influences more consumers compared to forms of electronic WOM, that however was not further investigated. Marketers are still reluctant to unveil the WOM potential and exploit the opportunities it includes, especially when it comes to new consumers that have no experience of the products. This study aims to shed light in the role of WOM for FMCG consumers in the pre-purchase stage, and investigate other factors that can possibly play also a role in this influence. The results reveal that indeed, tWOM has more influence to consumers
0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 1 2
0 (No/Small difference in Trust) 1 (Big difference in Trust)
Est im at ed Ma rgi nal Me ans Perceived Quality
37 | compared to eWOM touchpoints and that this influence is moderated by how trustworthy consumers perceive the different platforms the message is on.
5.1 Theoretical implications:
Results of this study provide plenty of answers to the theoretical gaps existing in the current literature. To begin with, this is one of the very few studies existing in literature concerning the FMCG sector and additionally the first experimental study on this sector. The findings of this research are applicable for consumers that are in the pre-purchase stage. Despite the great number of studies in WOM, authors were not investigating the pre-purchase stage of consumers. Hence, this study provides valuable theoretical insights concerning the consumers’ journey and WOM. This research revealed that consumers that have no previous experience of the product are influenced by tWOM, blogs and Twitter. This influence is translated in higher purchase intentions and higher perceived quality.
One research of Wu and Wang (2011), did try to measure the impact of WOM in fictional FMCG products but did not take into account specific touchpoints. Despite the fact the literature on WOM is of great extent there are not any studies existing that take into account the different WOM touchpoints in FMCG sector. Thus, another contribution to the existing literature is, the examination of the specific WOM touchpoints for the FMCG sector. This research reveals that, traditional WOM has a bigger influence in consumers than blogs or Twitter, which is aligned with the findings of Meuter, McCabe and Curran (2013) for the restaurant industry, even though different touchpoints concerning eWOM were investigated.
In addition, this study shed light to factors that are moderating this influence on consumers. Familiarity and Trustworthiness were tested, and results revealed that only perceived trustworthiness of the different touchpoints is moderating consumers perceptions. This result does not come as a surprise, however what is really interesting in the results is that share
38 | probability of the message follows a different pattern than purchase intention and perceive quality. The results show that share probability is not moderated by familiarity or trustworthiness, and additionally consumers intentions to share a message are similar across the eWOM touchpoints. This result adds more ground to the existing literature and also reveals further managerial implications.
Lastly, another interesting output of this study is that 36.4% of the respondents perceive a message from tWOM and eWOM as of equal weight. This percentage is expected to grow as we advance in the digital era. Messages are easily evaluated concerning their validity and truthfulness and also platforms are implementing mechanisms to prevent false messages from spreading.2 As a result consumers trust on digital technologies is increasing and it will not be a surprise to see the above percentage increasing in the upcoming years.
5.2 Managerial Implications
This research provides valuable insights to marketers and sheds light to positive WOM and its influence of consumers. To begin with, this study is focused on the pre-purchase level of consumers, hence the results may be of great importance when marketers are aiming to launch a new product, penetrate a new market or more simply attract new consumers from competition.
This study reveals that consumers’ perceptions are influenced by WOM. This influence is higher from tWOM, than from blogs and lastly Twitter, even though the last two are very close to each other. Therefore marketers should not be reluctant to use WOM marketing activities, and should not have doubt of their implications.
2 For related information you may visit The Washington Post and TechCrunch relative articles in the below links
(https://techcrunch.com/2012/02/15/facebook-verified-accounts-alternate-names/)
(https://www.washingtonpost.com/news/inspired-life/wp/2016/11/18/fake-news-on-facebook-is-a-real-problem-these-college-students-came-up-with-a-fix/?utm_term=.448b0759e6d1)
39 | Trustworthiness of the platform plays an important role on the influence of WOM. The more people trust specific platforms, blogs or individuals from which they are acquiring a message, the more influenced they are, concerning their purchase intention and the product perceived quality. An important figure that marketers should keep in mind is that 36.4% of people perceive tWOM and eWOM as the same. Despite this fact, marketers can also investigate the trustworthiness of the different platforms they use for campaigns in order to create more impactful activities in the future, by using more trustworthy platforms. However when the aim is dissemination of information, the touchpoints does not play an important role, as consumers are willing to share information no matter the source, if they find them interesting.
In general, given the insights provided by this study, practitioners are now able to create more effective WOM activities in the beginning of the customer journey. Depending on the purpose of the activities marketers can select those touchpoints that will be more effective. Furthermore, a possible trend is starting to unfold in which consumers start perceiving eWOM the same as tWOM, which is a fact that in the future will further assist marketers, as in eWOM there are more possibilities concerning the measurement and monitor of WOM.
5.3 Limitations & further research
This study despite the theoretical and managerial implications it provides, it also have certain limitations and indications for further research. To begin with, the most profound limitation of this study is that only positive WOM is investigated. Further research can provide with more insights concerning the negative WOM influence in FMCG consumers at the pre-purchase stage. Something relevant but only for Twitter and the box office industry was investigated by Hennig-Thurau, Wiertz and Feldhaus (2015). With the investigation of negative WOM academics and marketers will have a more holistic view on the matter.
